
Attackers Use Autonomous AI to Target Semiconductor, Defense and Civil Society Sectors
Our analysis of how attackers used the open-source Cairn AI agent framework to autonomously breach semiconductor, defense and civic targets.
35 research items
Journal articles, academic blogs, think-tank reports and papers in this section, newest first. Back to the news.

Our analysis of how attackers used the open-source Cairn AI agent framework to autonomously breach semiconductor, defense and civic targets.
Machine unlearning lets the owner of large language models (LLMs) remove specific entities from a trained model without retraining from scratch. Existing machine unlearning methods focus on removing the requested entities while maintaining utility. However, we observe that the representations of distinct entities in an LLM may overlap in shared parameter subspaces, a phenomenon known as knowledge superposition. Removing one entity may then affect other entities. This exposes…
Artificial intelligence agents increasingly automate cybersecurity tasks, including offensive operations. Frequent attacks across the web make endpoint detection and response (EDR) a critical defense, but whether AI agents can evade it remains unclear. Benchmarks evaluate the offensive capabilities of large language model (LLM) agents in areas such as penetration testing and vulnerability discovery. One important capability that these benchmarks do not address is defense…
What Congress should ask Defense Department officials about its use of AI for targeting and the protections in place against civilian harm.
The independent regulatory agency (IRA) model, a cornerstone of modern public administration since the Progressive Era, faces structural disruption as artificial intelligence (AI) and drone technology converge in environmental enforcement. This paper argues that AI and drone deployment do not merely augment agency operational capacity but fundamentally reconfigure the foundations of expertise, transparency, procedural due process, discretion, and accountability on which IRA…
Agentic AI systems using the Model Context Protocol (MCP) are vulnerable to indirect prompt injection through untrusted tool outputs. We demonstrate cross-modal trust laundering, in which instructions concealed in an email-signature image using least-significant-bit (LSB), discrete cosine transform (DCT), or metadata steganography bypass text scanners and enter a multi-agent workflow when decoded by a tool. To counter this threat, we propose decode-time provenance binding…
Artificial intelligence (AI) is increasingly used in military systems to support intelligence, surveillance, reconnaissance, autonomous operations, logistics, command and control, and decision support. Effective use, however, depends not only on technical performance but also on how operators perceive, interpret, evaluate, and rely on AI outputs under uncertainty, time pressure, and adversarial interference. This study presents a conceptual integrative review informed by a…
For eighty years, the Western alliance has anchored its strength in moral legitimacy, shared values, and the resulting credibility to shape international norms. NATO’s durability through the Cold War rested on more than nuclear deterrence alone: a shared conviction that the alliance represented a rules-based order worth sustaining, in which the use of force was bounded by law, constrained by democratic accountability, and subject to independent scrutiny. Dozens of nations…
Large language model (LLM) agents are moving from chat interfaces into infrastructure operations, where they read telemetry, call tools, generate and execute code, and change the state of production Kubernetes clusters. This collapses a boundary that conventional cloud-native security assumes: the boundary between data and control. Content that an agent merely reads (a log line, a ticket, a tool description) can redirect what it does. This paper argues that the model must…
This article examines the integration of artificial intelligence and autonomous weapons systems (AI/AWS) into Japan’s national security framework, exploring why a technologically advanced state facing severe regional threats has maintained a uniquely constrained development trajectory. While global military trends suggest a convergence towards AI-enabled capabilities, Japan’s adoption of these technologies is filtered through its strategic culture, which mediates how…
How does military artificial intelligence (AI) shape how war is known, understood, and acted upon? This article conceptualizes military AI as an epistemic machine – a system that mediates how the battlespace becomes intelligible. Drawing on the theory of technological mediation and critical AI scholarship, it examines how different AI systems mediate warfare through distinct epistemic lenses and military applications. Rather than treating AI as a singular technology, the…
Volume 68, Issue 5, October 2026, Page 43-58 .
In NATO’s multi-domain vision, artificial intelligence (AI) is considered the key to managing the enormous amount of data that is collected and exchanged. The article explores how military organizations can prepare for a future where AI permeates every level of operation – from administration to decision support and, ultimately, target engagement. The analyses focus on the practical application of AI within the US military prior to the second Trump Administration. Drawing on…
The increasing number of artificial intelligence (AI) elements within military systems has introduced new forms of security dilemmas related to speed, level of secrecy, and transfer of responsibility from humans to machines. This article addresses the question of whether or how AI enabled autonomous weapons can lead to greater levels of strategic instability. Three causal mechanisms were determined by this study to potentially create destabilizing effects due to the…
Why do fragmented governance arrangements sometimes become organised without producing hierarchy or substantive agreement? This article answers that question through the relationship between the Responsible AI in the Military Domain (REAIM) process and the United Nations General Assembly’s agenda on military AI. Drawing on primary and secondary sources, including elite interviews with diplomats serving at UN missions in New York and Geneva, it traces how REAIM widened…
The Model Context Protocol (MCP) is an open-source standard that allows AI agents to connect to external tools, databases, and services. While this connectivity enables powerful agent capabilities, it also introduces multi-step attacks that existing per-call defenses cannot reliably detect. Attackers can compose individually benign tool invocations into malicious sequences that evade isolated inspection. This paper presents ChainWatch, a sequential detection framework for…
Abstract Artificial Intelligence (AI) systems are increasingly supporting targeting, intelligence analysis and operational planning across military domains, reshaping how commanders use force through human–machine teaming (HMT). HMT offers operational advantages, but risks such as automation bias, adversarial manipulation and degraded performance pose challenges for the ability of deployers to use AI systems in compliance with international humanitarian law (IHL). This…
Debates on military artificial intelligence (AI) remain skewed toward great power dynamics, oscillating between techno-optimist celebrations of speed and techno-pessimist warnings of collapse. This binary overlooks how middle powers, operating under nuclearized rivalry and asymmetric sanction risk, embed restraint into organizational and technical practice. This article develops Systems Restrained Realism (SRR), a framework that extends defensive realism into the machine age…
In the context of unprecedented U.S. Department of Defense (DoD) budgets, this paper examines the recent history of DoD funding for academic research in algorithmically based warfighting. We draw from a corpus of DoD grant solicitations from 2007 to 2023, focusing on those addressed to researchers in the field of artificial intelligence (AI). Considering the implications of DoD funding for academic research, the paper proceeds through three analytic sections. In the first…
This article examines how artificial intelligence systems deployed in military operations generate ‘algorithmic psychopolitics’ which is a transformation of warfare wherein automated decision architectures operate beyond human moral comprehension. Drawing on Byung-Chul Han’s philosophy and news report analysis of the Israel Defense Forces’ AI targeting systems, particularly the Lavender program, this investigation shows how military AI establishes a ‘digital unconscious’…
Deploying large language models (LLMs) as autonomous browser agents exposes a significant attack surface in the form of Indirect Prompt Injection (IPI). Cloud-based defenses can provide strong semantic analysis, but they introduce latency and raise privacy concerns. We present the Cognitive Firewall, a three-stage split-compute architecture that distributes security checks across the client and the cloud. The system consists of a local visual Sentinel, a cloud-based Deep…
Discourses on nuclear weapons and military applications of artificial intelligence (AI) portray them either as apocalyptic super weapons, posing catastrophic risks, or as indispensable to states’ national survival and the international security architecture. At the same time, debates about and governance efforts to regulate these weapons have become contested, stalled, or even abandoned. We examine the intersection of both regulatory discourses by asking: How do contemporary…
This article critically examines the control-by-design imaginary in the development of autonomous weapons systems (AWS): the belief that technical safeguards and human ingenuity can resolve the risks they pose. Adopting a technopolitical lens, it conceptualizes AWS design as a heterogeneous process shaped by technical, institutional, and political forces. The analysis highlights two key practices: encoding target profiles and determining and validating error rates. It argues…
As AI shifts from human-in-the-loop interfaces to autonomous multi-agent systems capable of real-time code execution and tool integration through protocols like the Model Context Protocol (MCP), traditional SAST, DAST, and legacy AI safety methods fail to detect modern agentic-AI threats. This paper introduces the LLM Scalability Risk Index (LSRI), a parametric framework and cybersecurity standard for stress-testing autonomous orchestration pipelines. LSRI measures the…
Volume 35, Issue 162, November 2026, Page 2155-2169 .
Read at Journal of Contemporary China
Also at Journal articles (OpenAlex)
This article analyses military AI governance using the Advocacy Coalition Framework (ACF), focusing on two key advocacy coalitions: the ‘tight regulation’ coalition which emerged in the context of the UN’s Group of Governmental Experts on Lethal Autonomous Weapons Systems, and ‘responsible use’ coalition which emerged in the context of the REAIM Summits and the US-led Political Declaration on Responsible Military AI Use. The ‘tight regulation’ coalition advocates for a…
ABSTRACT This study examines how the use of decision‐support military Artificial Intelligence (AI) systems can affect the democratic conduct of warfare. AI can challenge the democratic conduct of warfare by introducing systemic risks such as reduced oversight, opacity, and automation bias. The inquiry focuses on two case studies: US Project Maven and Israeli AI targeting systems. Available reporting and official statements portray the United States as emphasizing procedural…
The study examines Pakistan's artificial intelligence (AI) strategy for national security and defence. It identifies AI trends and effects using qualitative data from 105 military, intelligence, cybersecurity, and government specialists. Thematic research indicates AI improves military decision-making, intelligence synthesis, surveillance, and real-time risk assessment. Participants emphasised the importance of AI in military resource allocation, target precision, predictive…